I've offered to help a friend with a short film project that he's producing based on the after-effects of Nuclear fallout, and he requires me to create a mushroom cloud/nuclear explosion.

I've been playing with Pyro in my sparetime to approximate this, mainly with the fireball tool and keyframing several parameters to get a high temperature, fast moving initial fireball that is followed by a cooler ‘chamber’ of smoke, and while my results are looking okay, I was wondering if anyone had any other suggested workflows for how I could go about attempting this. I've seen some very complex solutions over on OD-force (which seems to be down as I write this) however these seem to be more focused on the static post-explosion mushroom cloud.

I'm mainly concerned that I'm missing a very obvious method or workflow to achieve this effect by sticking quite rigidly to the pyro preset. I have very little experience with the various fluid/gas solvers and subsolvers (although I would really love to begin to gain a basic understanding of these seemingly very powerful tools) however perhaps these could be used to gain more control over the various aspects of a mushroom cloud.

My other idea was that I could draw out the profile of a mushroom cloud with a nurbs curve, use a vop-sop to rotate the normals to follow the direction of a curve and then somehow pipe this into the simulation to control the movement of the fluids.

Here is an example that my friend showed me of what he would like me to achieve:
http://www.youtube.com/watch?v=zsTRxXvQY0s [youtube.com]

I think I could achieve this in Pyro, however as I say I would be very interested in hearing if anyone had any suggestions.

Many thanks,
Gary

I've found using particles as a fuel source for any fast moving explosions works very well. Start by creating a particle simulation with a very quick initial burst and short life. You might also consider colliding some with a ground plane and have them stick so that they constantly emit fuel from the ground (this will help with the smoke column). Then, use that simulation as fuel for your pyro sim - be sure to inherit the particles velocity (an option in the pyro solver). You're correct that you will need to key many parameters such as temperature, turbulence, and buoyancy lift over time to achieve realistic results. Your idea of drawing a curve to control the fluids sounds interesting, but may not be necessary as I find that pyro explosions love to turn into mushroom clouds all on their own

kacheson
I find that pyro explosions love to turn into mushroom clouds all on their own

raising your Vorticle Confinement in the pyro solver helps to accentuate the the swirling motion of fluids and helps it swirl more in on itself. I would also mess with keyframing the burn rate. The more fuel you burn, the bigger of an explosion you can hope to achieve.

Thanks for the replies guys, it's given me food for thought.

Kach, do you bake your particle sim out to geometry and then bring that into the Pyro sim as deforming static geometry, or do you somehow use the particles to generate fuel? I ask as I've tried using a Particle Location and POPsolver Dop in the pyro sim, but it doesn't seem to be generating any fuel based on this.

I've attached the file if you wouldn't mind having a quick look at it.

Many thanks,
Gary

My mistake, I let the sim run for longer and it looks like they are emitting fuel, it just takes a while!

Sorry about that.

They should emit fuel immediately, and lots of it. Also, I failed to mention that you must make sure you've added some temperature under the ‘Physical’ tab of the pop object. Your fuel won't burn if this is set to 0. Obviously the higher you set this value, the greater it will effect your simulation with regards to buoyancy lift, etc.

Ponge
I've offered to help a friend with a short film project that he's producing based on the after-effects of Nuclear fallout, and he requires me to create a mushroom cloud/nuclear explosion.

I've been playing with Pyro in my sparetime to approximate this, mainly with the fireball tool and keyframing several parameters to get a high temperature, fast moving initial fireball that is followed by a cooler ‘chamber’ of smoke, and while my results are looking okay, I was wondering if anyone had any other suggested workflows for how I could go about attempting this. I've seen some very complex solutions over on OD-force (which seems to be down as I write this) however these seem to be more focused on the static post-explosion mushroom cloud.

I'm mainly concerned that I'm missing a very obvious method or workflow to achieve this effect by sticking quite rigidly to the pyro preset. I have very little experience with the various fluid/gas solvers and subsolvers (although I would really love to begin to gain a basic understanding of these seemingly very powerful tools) however perhaps these could be used to gain more control over the various aspects of a mushroom cloud.

My other idea was that I could draw out the profile of a mushroom cloud with a nurbs curve, use a vop-sop to rotate the normals to follow the direction of a curve and then somehow pipe this into the simulation to control the movement of the fluids.

Here is an example that my friend showed me of what he would like me to achieve:
http://www.youtube.com/watch?v=zsTRxXvQY0s [youtube.com]

I think I could achieve this in Pyro, however as I say I would be very interested in hearing if anyone had any suggestions.

Many thanks,
Gary

Thanks you for the post.